Taking a holistic approach to a revamp

24 Mar.,2023


If all aspects of a system are carefully examined, even minor changes can lead to large gains in unit performance and profitability.

Article Summary

Engineers frequently focus on one piece of the puzzle as a cause for poor performance. Blinkers on, fixing that one problem becomes the project goal. A tray specialist will focus on the tray design, a hydraulic engineer will focus on hydraulics, and so on. While some improvement is obtained, many opportunities to yield major benefits can be missed. Instead, a holistic review of the entire plant operation by engineers familiar with all disciplines best identifies and captures the processing opportunities.
A recent review of a butane splitter illustrates how analysing the entire system — trays, reboiler, condenser, controls, and operation — uncovered multiple changes and modifications that, if implemented individually, would each improve Isobutane recovery to varying degrees.

Taken together, however, finding and implementing all these changes resulted in a predicted 50% increase in isobutane recovered at design charge rates. A review of post- revamp operation has shown that the modified tower has improved the refinery’s bottom line by at least $2.6 million/y through a combination of improved operating guidelines and capital fixes to existing design issues.

The butane splitter column presented here recovers isobutane from a mixed C4 stream originating in the crude unit saturates gas plant (SGP). The isobutane is used as make-up to the alkylation unit.

The tower was originally built in 1965 for a different service and was converted to butane splitter service later. Diameter limited, the column had been revamped in 1999 with high capacity trays and some additional condensing capacity. The tower had under-performed since that revamp, leaving excess isobutane in the bottom product. This required the refinery to purchase additional isobutane for the alkylation unit and limited the amount of butane that could be blended into gasoline, both of which resulted in a considerable economic penalty.

A study was commissioned a few years after the 1999 revamp to determine the causes of poor separation. Since the trays were changed from conventional valve trays to high capacity trays during the revamp, it was felt that the trays were at fault, and the study focused on the tray design and efficiency. The study presented several recommendations regarding tray design details and proposed the tower be retrayed again. No action was taken at that time.

Several years later, a short column outage was planned that would allow potential modifications to improve tower performance. Ascent Engineering was asked to review the previous study and other, more recent troubleshooting efforts and propose modifications that would improve the tower performance. Based on the previous study, new trays were an expected recommendation. Reboiler performance was a known issue. The upcoming shutdown was expected to be short, and the time allotted would not allow for the installation of new major equipment such as a new reboiler. Tie-ins for new equipment could be made if necessary. Because the shutdown had already been scheduled prior to the project kick-off, the new study was fast tracked in order to meet the already planned shutdown window.

The project methodology Ascent used was not only to focus on the tower trays, as was previously done, but to take a holistic approach to look at the entire system. This was particularly important since the tower had been designed for a different service, and some aspects of the equipment layout may not have been ideally suited for butane splitter service. Ascent started with a very detailed plant match simulation and evaluation of all major system components to identify deficiencies and opportunities. Verifying the reboiler, the condenser, the controls, the operating philosophy, and the other tower internals in addition to the trays was required if the project was to be a success.

Butane splitter operation
As Figure 1 shows, a mixed C4 stream from the SGP is preheated by a set of feed/bottoms exchangers before entering the butane splitter between trays 35 and 34. The tower overhead vapour is condensed to bubble point in a bank of air coolers. There are eight parallel air cooler bays in this bank.

The bottom tray is a chimney tray which collects liquid and gravity-feeds an elevated kettle reboiler. The reboiler vapour return is distributed by a Schoepentoeter, located just above the chimney tray. Schoepentoeter is a proprietary Shell vane type inlet device that is used to introduce gas/liquid mixtures into a vessel or column. The liquid from the reboiler gravity flows to the tower sump.

The butane splitter feed is fractionated into two products:
- An isobutane distillate which provides part of the alkylation unit’s isobutane make-up requirements.
- Normal butane and heavier material from the column bottom which is pressurised out to a storage sphere and eventually blended to gasoline.

Column operation was reported to be steady. Previous tower scans showed no evidence of flooding. The tower appeared to be reboiler limited since increasing steam flow beyond a certain point resulted in condensate flow meter instability and control valve swings. The amount of isobutane in the bottoms was significantly more than the 1999 design prediction, even at charge rates well below design. Because the tower pressure had to be set high enough to allow the bottom product to be pressured out to storage (approximately 100 psig), the tower seemed to have excess condensing capacity.

Ascent collected data sheets and mechanical details for all equipment in the system. The client provided operating data from the DCS system and product analyses from the laboratory.

Initial calculations and test run
With this information in hand, we proceeded to evaluate all aspects of the system. Tower operation was simulated and rigorously verified against the operating data. Tray and tower drawings were closely reviewed, and flooding calculations were performed for the trays. The tower control system and logic was reviewed. The reboiler and condensers were rigorously rated. A nozzle elevation review revealed that there was little static head driving the reboiler flow, so detailed hydraulic calculations for the reboiler circuit were performed.

The initial simulation results and equipment calculations showed that the tower’s performance with existing equipment was much worse than calculated. The primary issues identified were hydraulic limitations with the reboiler piping and issues with the control scheme and operating philosophy. Tray flooding was found to be unlikely, though low tray efficiency (which had been previously identified by the client) was confirmed to be a problem.

A plant test was scheduled to determine whether changes in operation could improve tower performance immediately. Gamma scans of the tower were also scheduled to see if the trays were prematurely flooding and to help confirm the conclusions made regarding the reboiler piping having insufficient head. The results of these tests were used to develop the final scope of modifications recommended to improve tower performance, maximise isobutane recovery, and maximise refinery profit